核心概念
Reinforcement learning optimizes battery management in dairy farming, reducing costs and grid reliance.
摘要
This study explores using Q-learning to manage battery charging and discharging in dairy farms. The research aims to reduce electricity costs and peak demand by optimizing renewable energy utilization. By integrating wind generation data, the algorithm achieved a 24.49% reduction in imported electricity cost. Expanding the state space improved performance, but challenges arose due to dimensionality. Testing on an Irish dataset showed a 6.7% reduction in electricity imports compared to the baseline method.
統計資料
The proposed algorithm reduces the cost of imported electricity from the grid by 13.41%
Peak demand reduced by 2%
Wind integration led to a 24.49% reduction in electricity costs
Q-learning achieved a 6.7% reduction in electricity imports on the Irish dataset